Libraly Journal

Libraly Journal ›› 2025, Vol. 44 ›› Issue (414): 13-22.

Previous Articles     Next Articles

The Integration Models of Generative Artificial Intelligence inInternational Advanced Platforms for S&T Information and Communication

Xu Hongyu (Shanghai Library)   

  • Online:2025-10-15 Published:2025-11-06
  • About author:Xu Hongyu Shanghai Library

Abstract:

This paper presents a systematic study of the integration of generative AI into advancedplatforms for S&T information and communication in the international context. It analyzes and comparesthe AI application pathways and development characteristics of both traditional and emerging platforms.The paper first distinguishes between the structural ecologies of the two types of platforms traditionalplatforms—including major academic publishers professional information service providers and publicinstitutions—are characterized by strong content authority and complete service chains emergingplatforms such as preprint servers open data repositories and unofficial literature platforms emphasizeopenness sharing and collaborative innovation. Based on this distinction the paper identifies threetypical AI integration models among traditional platforms vertically self-developed AI strategicpartnerships with large external models and content licensing collaborations highlighting a trend towardcentralized management and commercial orientation. For emerging platforms it outlines three AIintegration pathways data and interface openness collaborative crowdsourcing and embedded selfdevelopedtools which reflect an ecosystem-driven and developer-oriented logic. Finally throughcomparative analysis the paper concludes that the two types of platforms exhibit distinct integrationlogics the centralized and closed-loop model versus the ecosystem-oriented and open model. Thesefindings provide both practical insights and theoretical references for the future development of advancedcommunication platforms.

Key words:

Advanced platforms for S&, T information and communication Generative artificial intelligenceAI integration pathways Open science Academic publishing transformation